Underestimation of the Presence of Breast Carcinoma in Papillary Lesions Initially Diagnosed at Core-Needle Biopsy

Abstract

To retrospectively determine the degree of underestimation of breast carcinoma diagnosis in papillary lesions initially diagnosed at core-needle biopsy. Institutional review board approval and waiver of informed consent were obtained for this HIPAA-compliant study. Mammographic database review (1994-2003) revealed core biopsy diagnoses of benign papilloma (n=38), atypical papilloma (n=15), sclerotic papilloma (n=6), and micropapilloma (n=4) in 57 women (mean age, 57 years). Excisional or mammographic follow-up (>or=2 years) findings were available. Patients with in situ or invasive cancer in the same breast or patients without follow-up were excluded. Findings were collected from mammography, ultrasonography, core technique, core biopsy, excision, and subsequent mammography. Reference standard was excisional findings or follow-up mammogram with no change at 2 years. Associations were examined with regression methods. In 38 of 63 lesions, surgical excision was performed; in 25 additional lesions (considered benign), follow-up mammography (24-month minimum) was performed, with no interval change. In 15 lesions, 14-gauge core needle was used; in 48, vacuum assistance (mean cores per lesion, 8.7). Carcinoma was found at excision in 14 of 38 lesions. Core pathologic findings associated with malignancy were benign papilloma (n=1), sclerotic papilloma (n=1), micropapilloma (n=2), and atypical papilloma (n=10). Frequency of malignancy was one (3%) of 38 benign papillomas, 10 (67%) of 15 atypical papillomas, two (50%) of four micropapillomas, and one (17%) of six sclerotic papillomas. Excisional findings included lobular carcinoma in situ (n=2), ductal carcinoma in situ (n=7), papillary carcinoma (n=2), and invasive ductal carcinoma (n=3). Low-risk group (micropapillomas and sclerotic and benign papillomas) was compared with high-risk atypical papilloma group. Core findings were associated with malignancy at excision for atypical papilloma (P=.006). Lesion location, mammographic finding, core number, or needle type were not associated (P>.05) with underestimation of malignancy at excision. Benign papilloma diagnosed at core biopsy is infrequently (3%) associated with malignancy; mammographic follow-up is reasonable. Because of the high association with malignancy (67%), diagnosis of atypical papilloma at core biopsy should prompt excision for definitive diagnosis.